5 datasets found

Ref-DAVIS17 is an extension of the DAVIS17 dataset, where it enhances the dataset by providing language descriptions for each specific object present in the videos.

Referring video object segmentation (RVOS) aims to accurately segment the target object in the video with the guidance of given language expressions.

G-Ref is a dataset for referring image segmentation, comprising 104K referring language expressions for around 55K objects in about 27K images.

Ref-Youtube-VOS is an extensive referring video object segmentation dataset that comprises approximately 15,000 referring expressions associated with more than 3,900 videos.

The dataset used in the paper is a benchmark for referring expression grounding, containing 142,210 referring expressions for 50,000 referents in 19,994 images.